0.12.0 Beta 3 #5039
Replies: 21 comments 64 replies
-
I have created a short survey for those that are using the new OpenVINO and TensorRT detectors. It will aid in understanding the expected performance for different devices and help identify other issues that may be occurring. Thanks! Answers can bee seen here |
Beta Was this translation helpful? Give feedback.
-
Thanks very much for another great update and for your hard work. I've just loaded it up and it seems hw acceleration is now working on my Pi 4 as I now get a video output with it enabled! Super nice! I can log an issue but I'm sure I'm just doing something stupid again. I was getting a 404 could not be found error for both my cameras. Example: So I tried specifying the IP specifically and it fails with a different error.
Live views all work fine but there are no recordings. Perhaps it's a portainer issue but recordings were working fine in the previous beta. I will continue troubleshooting. |
Beta Was this translation helpful? Give feedback.
-
Unraid template updated, will be avaible on the store as a new app "frigate12-beta3" |
Beta Was this translation helpful? Give feedback.
-
Let me know if I shld create a separate issue.... I'm testing on a spare Pi3b while waiting for Pi4s to come back in stock. On the latest beta3. Everything works fine but when I add hwaccel_args, I get the colored screen instead of my cam feed. Removing it, everything works.
|
Beta Was this translation helpful? Give feedback.
-
Updated to beta 3, with intel hw accel I still have cameras running to a higher FPS that configured (5): The CPU usage is still higher than usual. With tensorrt, all the cameras are running to 5FPS and CPU usage looks normal: |
Beta Was this translation helpful? Give feedback.
-
I just thought I'd share the following two things I've found useful: If you wish to use hardware accel with h264 and h265 on the Raspberry Pi 4 use:
If you are having any trouble with recording or are seeing 404 not found errors in your logs try adding the following to your config:
Thanks to Nick for helping me with this issue and finding the resolution :). |
Beta Was this translation helpful? Give feedback.
-
Beta Was this translation helpful? Give feedback.
-
Running Frigate stand-a-lone (not in the HA AddOn ecosystem). From what I can see beta3 has not been published to hub.docker.com. https://hub.docker.com/r/blakeblackshear/frigate/tags Can someone please fix? :) Thanks. |
Beta Was this translation helpful? Give feedback.
-
I was on 0.11 and updated to 0.12-beta3. I only have one camera enabled (I was/am in the test phase). Everything works fine until I add an input with a "record" role, for example: - path: rtsp://X
roles:
- record In which case I start getting this error in the logs: "Camera doorbell has restream enabled, but restream is not assigned to an input.", and the app refuses to start (seems like it should allow startup in more situations now that config can be edited in the UI). Here is my full config file: mqtt:
host: 192.168.x
user: x
password: x
detectors:
coral:
type: edgetpu
device: pci:0
ffmpeg:
hwaccel_args: preset-vaapi
cameras:
doorbell:
ffmpeg:
inputs:
- path: rtsp://X
roles:
- detect
- path: rtsp://X
roles:
- record
output_args:
record: preset-record-ubiquiti
rtmp: preset-rtmp-ubiquiti
detect:
width: 960
height: 720
rtmp:
enabled: False # <-- RTMP should be disabled if your stream is not H264
record:
enabled: True
retain:
days: 7
mode: motion
events:
retain:
default: 14
mode: active_objects
objects:
dog: 2
car: 7 |
Beta Was this translation helpful? Give feedback.
-
Finding a couple issues related to webrtc, recording, and transcoding h265 to h264. Camera is streaming with hevc codec, if restream is enabled and we copy the video_encoding, and enable recording, then the WebRTC stream only has audio - no video (in safari and Chrome 108+) but the recording seems to work just fine. Config below:
If we transcode to h264 in the restream, the WebRTC stream works fine, but recordings continuously fail. Only video_encoding was changed in restream as shown:
Looks to me like it's expecting hevc but the stream has been transcoded so it's not expecting that h264 codec
|
Beta Was this translation helpful? Give feedback.
-
Hey guys Im back again. Im getting the same 98 error Address already in use. I went ahead and reinstalled ubuntu server and then CasaOS on top https://casaos.io/ After getting ubuntu and casaos installed, the only docker I installed first was frigate nvr....it wouldn't load. Then I installed portainer, which showed I was getting the same error. There's nothing that shows that would be already in use. Is there something else Im not seeing?
|
Beta Was this translation helpful? Give feedback.
-
Man, WebRTC is incredible and this beta is amazing. Thank you! |
Beta Was this translation helpful? Give feedback.
-
Hello, I have an issue with hwaccel. I have fanless computer with N5105 processor. If I try vaapi or qsv it crashes, it runs ok without hwaccel but high cpu usage. I tried with environnement vars i965 as well. libva: /usr/lib/x86_64-linux-gnu/dri/iHD_drv_video.so init failed maybe I can produce some useful logs ? |
Beta Was this translation helpful? Give feedback.
-
I'm just wondering, what exactly does the Apart from that, can I also use this feature to enable audio in recordings? Like: cameras:
video_porteiro:
ffmpeg:
inputs:
- path: rtsp://192.168.1.40/cam/realmonitor?channel=1&subtype=0&unicast=true&proto=Onvif
roles:
- restream
- path: rtsp://127.0.0.1:8554/video_porteiro?video=copy
roles:
- detect
- path: rtsp://127.0.0.1:8554/video_porteiro?video=copy&audio=aac
roles:
- record
output_args:
record: preset-record-generic-audio
restream:
force_audio: true
detect:
width: 1280
height: 720 I'm assuming that audio in recordings is still preferred to be in AAC format. Notice that I'm using |
Beta Was this translation helpful? Give feedback.
-
I just realized that the playback is choppy when I use the restream to record with Reolink. The playback is smooth when I use the
|
Beta Was this translation helpful? Give feedback.
-
Just adding how happy I am with this beta. Massive improvements from 11. Preset nvidia acceleration works great, dropped CPU usage substantially and surprisingly even my GPU utilization dropped a bunch. And I think this version might actually get wife approval with the new WebRTC stream quality/speed. As if Frigate wasn't already awesome enough, this version takes it to a new level. Keep up the awesome work. |
Beta Was this translation helpful? Give feedback.
-
Tried to run tensorrt image with nvidia gpu, crashed with missing library 'libnvrtc'
Tried to install library top of beta3 image:
It seems to run fine now |
Beta Was this translation helpful? Give feedback.
-
What's the |
Beta Was this translation helpful? Give feedback.
-
Am I the only one who had issues because I had use_experimental set as true? I couldn't access the pages due to javascript errors until I turned it off. Maybe a note would help in the breaking changes list for 12? Maybe just me though :) |
Beta Was this translation helpful? Give feedback.
-
Does anyone know how to make models for |
Beta Was this translation helpful? Give feedback.
-
with this release what is the correct setup for restream/go2rtc? i have this currently and am getting an error for extra fields not permitted on my go2rtc section. here are my relevant config sections:
|
Beta Was this translation helpful? Give feedback.
-
Major Changes for 0.12.0
❗❗Notice❗❗
I want to recognize @NickM-27 for all the contributions he made on this release and all the support he helps provide in the issues. If you have been considering sponsoring this project with either a one time contribution or a recurring contribution, I would request that you do so at his sponsors page.
Link to updated docs: https://deploy-preview-4055--frigate-docs.netlify.app/
Changes Since Beta 2
ghcr.io/
in front of your image names. (blakeblackshear/frigate:0.12.0-beta3
needs to be changed toghcr.io/blakeblackshear/frigate:0.12.0-beta3
)BREAKING CHANGES
./run: line 7: exec: nginx: not found
, then you need to stop overriding your PATH. This happens with both Proxmox LXC and Portainer ([Support]: Beta11 Can't Find FFMPEG #3223 the database will need to be deleted to roll-back successfully.live
configuration has been moved torestream -> jsmpeg
Retention limited by available storage
Frigate now limits its recordings to available storage, if the storage for recordings gets below 1 hour left then Frigate will delete the oldest recording segments to make room for newer ones. Frigate will also not fail if there is no space to save recordings.
RTSP restreaming
Due to many issues with RTMP based restream, RTMP is now deprecated in favor of a new RTSP restream. This also comes with a new
restream
role which is enabled by default. The restream of a camera can be accessed atrtsp://frigate_ip:8554/config_camera_name
. The restream can also be used inside Frigate to reduce connections to the camera, see the restream docs for more info.Birdseye restreaming is also now supported at
rtsp://frigate_ip:8554/birdseye
NOTE: Port
8554
will need to be mapped in the docker run or docker compose file for the restream to be accessed outside the container.New live stream options
The new RTSP restream also enables new
MSE
andWebRTC
live view options in the frontend which enable low-latency full-framerate live views that support audio.NOTES:
The live view options are set in the Frigate WebUI for each camera individually.
New detector types
Frigate now supports new detector types along with the Google Coral TPU.
OpenVINO
The OpenVINO detector type runs an OpenVINO IR model on Intel CPU, GPU and VPU hardware. OpenVINO is supported on 6th Gen Intel platforms (Skylake) and newer. A supported Intel platform is required to use the GPU device with OpenVINO. The MYRIAD device may be run on any platform, including Arm devices. For detailed system requirements, see OpenVINO System Requirements
TensorRT
NVidia GPUs may be used for object detection using the TensorRT libraries. Due to the size of the additional libraries, this detector is only provided in images with the -tensorrt tag suffix. The TensortRT detector is able to run on x86 hosts that have an Nvidia GPU which supports the 11.x series of CUDA libraries. The minimum driver version on the host system must be >=450.80.02. Also the GPU must support a Compute Capability of 5.0 or greater. This generally correlates to a Maxwell-era GPU or newer, check the TensorRT docs for more info.
NOTE: The link in the docs is for the final release location, for the beta the script is located at https://raw.githubusercontent.com/blakeblackshear/frigate/dev/docker/tensorrt_models.sh
New system page
The
Debug
page has been renamed toSystem
. It now includes much more information and links to easily getvainfo
andffprobe
data.The GPUs section will show each GPU based on the hwaccel args, depending on the type of GPU different information is supported. For example Nvidia GPUs will have the name while others will be generic, also intel does not support memory usage. There is also a
VAINFO
button which will make it easier to diagnose hwaccel issues and verify that the hwaccel driver is being used correctly.Each camera will have their own process CPU & memory usage so it will be easier to see which process is using those resources. There is also an
FFPROBE
button which will be helpful to understand what each stream is presenting and make it easier to include that in support issues.The debug config has also been removed, the config can be copied using the config page in the WebUI.
New storage page
Frigate WebUI now has a built in storage page which shows the general usage for both storage and memory. It also shows a per-camera storage usage and stream bandwidth. This should make it a lot easier to have a general idea how much storage is being used by each camera.
New logs page
Frigate WebUI now has built in logs, this will make copying and viewing logs much easier. Currently the logs are static meaning the page must be refreshed to view new logs.
New config editor
Frigate WebUI now has a built in config editor with syntax highlighting and schema validation. This means that the config will be validated before it can be saved which will make yaml config much easier to manage.
NOTE: The previous docker-compose or docker CLI may have had the /config.yml set to
:ro
or read-only. That will need to be removed for the config editor to be able to save the new file.MQTT is now optional
Frigate no longer requires MQTT to function, MQTT is also setup asynchronously so errors will be more clear and frigate won't stop when mqtt is enabled but not setup correctly.
NOTE: MQTT is still required for the Frigate-HomeAssistant Integration
No changes are needed for existing users.
FFMpeg presets
FFMPEG presets for common configurations are now added, making the config cleaner and allowing the underlying args to be changed between releases without being a breaking change. It is highly recommended to update the configuration to use these presets.
See https://deploy-preview-4055--frigate-docs.netlify.app/configuration/ffmpeg_presets for more info on how these should be used.
Example: Enabling Audio In Recordings
before:
after:
This also enables presets for hardware acceleration which will be used to further reduce CPU usage by using the GPU to optimize scaling and other workloads.
Improved events filters
Recordings updates
Recording segments are now stored in UTC to avoid issues with DST. In order to avoid a breaking change, the path of the stored segments has changed from
/media/frigate/recordings/%Y-%m/%d/%H-%s.mp4
to/media/frigate/recordings/%Y-%m-%d/%H-%s.mp4
.Recordings in the frontend are now shown in the timezone of the device viewing frigate, so the timezone set on the server running frigate is irrelevant.
No changes are needed for existing users.
What's Changed
-user_agent
forrtmp
streams by @felipecrs in Do not apply-user_agent
forrtmp
streams #4596New Contributors
Full Changelog: v0.11.1...v0.12.0-beta1
Images:
ghcr.io/blakeblackshear/frigate:0.12.0-beta3
ghcr.io/blakeblackshear/frigate:0.12.0-beta3-tensorrt
This discussion was created from the release 0.12.0 Beta 3.
Beta Was this translation helpful? Give feedback.
All reactions